> What is the most accepted explanation for why we have the independent and
> dependent variables reversed in supply and demand graphs?  It came up in
> class today and I gave “an” answer, but I admitted to my class that I wasn’t
> confident that I was correct.  I also promised them I’d ask all of you.
>

Here's my answer, a bit long, but I think it does completely answer
the question.

It comes from my micro book (Intermediate Microeconomics with
Microsoft Excel, CUP, 2009).

I chose as an epigraph for the chapter on deriving demand, the following:

"The first “empirical” demand schedule was published in 1699 by
Charles Davenant.
George J. Stigler"

Then, in the references to the chapter, I cite and explain:

"The epigraph is from page 103 of George J. Stigler, “The Early
History of Empirical Studies of Consumer Behavior,” The Journal of
Political Economy, Vol. 62, No. 2. (Apr., 1954), pp. 95-113.

Most economists do not care who first came up with the concept of a
demand schedule.  Most of those who do care believe that it was
Gregory King, a century after Charles Davenant.  Stigler was a winner
of the Nobel Prize in Economics and a professor at the University of
Chicago.  He had a lifelong passion for the intellectual history of
economics.  In this article, he showed that Davenant actually preceded
King.

It took a long time to translate demand (and supply) schedules as
tables (with columns for price and quantity) into graphs.  Although
there were precursors, Alfred Marshall’s Principles of Economics
(1890) is credited with introducing supply and demand graphs to
English speaking economists.  These graphs appeared, however, only in
footnotes.

Marshall’s Principles was the most popular economics book of its era.
It is freely available online at
<www.econlib.org/library/Marshall/marP.html>.

Marshall put price on the vertical axis because he wanted to show
market demand and supply curves on a graph as the horizontal sum of
individual demand and supply curves, as in footnote 70 from Book III,
Chapter IV.  Future generations of introductory economics students
became locked in to the Marshallian inverse demand and supply curves.

Although you may conclude that Marshall’s violation of accepted
mathematical convention (i.e., independent variables belong on the x
axis) is confusing, the decision was not based on ignorance.  In fact,
Marshall was a brilliant mathematician, earning Second Wrangler (to
the future Lord Rayleigh) as an undergraduate at Cambridge in the
Tripos competition."


FYI, I added the last paragraph because I think many economists
believe that P on the y axis is some kind of stupid mathematical
mistake. I hope we can all agree that is not correct. There is an even
longer explanation based on Marshall's equilibration process, but
maybe others will focus on that.

-- 
Humberto Barreto
www.depauw.edu/learn/microexcel